This project focuses on the executive institutions of judicial appointments and legal advocacy within the federal judicial system. Examination of prosecutorial decisions and case outcomes across federal districts, gathered from TRAC data, will be used to describe the extent and conditions of presidential influence within the judiciary. Whether, how, and when the executive moves legal policy towards his most preferred outcome is the objective of this research program.

How does the organizational structure of a bureaucracy impact the extent of corruption? I also rely on an objective corruption measure, the data on the prosecution of federal corruption cases by US federal attorneys. I also use aggregated American federal personnel records on approximately 900,000 employees annually for 24 years to reconstitute the structure of each local branch of the US federal administration. I investigate four hypotheses related to network analysis: (1) are branches that are characterized by denser social networks less prone to corruption? (2) Are branches that are characterized by more strong ties more prone to corruption? (3) Are branches that are characterized by flatter hierarchies less prone to corruption? (4) Do increases in the probability of detection result into larger corruption rings?

We use variation in corruption convictions due to changes in state-level enforcement policies to measure the impact that public corruption has on administrative costs in state governments. We think about state legislatures as being primarily responsible for delivering services to state residents. The services delivered come at some administrative cost. Corruption has the potential to either increase the costs of administration or decrease output, either of which can result in reduced efficiency. This project's primary aim is to estimate the magnitude of the reduction to efficiency and investigate potential policy solutions.

I estimate the average impact that a newspaper has on public corruption convictions at the local, state, and federal levels in the United States. I am able to identify the causal effect by using newspaper entry and exit decisions, which are unlikely to be caused by or be highly correlated with omitted variables affecting corruption convictions (after controlling for state and year fixed effects). Furthermore, assuming that the stock of corrupt officials is slower to adjust following a subtle policy change than the probability of detection, I can attribute most of the observed impact to changes in detection itself, holding corruption levels constant.

This paper investigates the spatial dynamics of shadow economies, corruption, and the corruption-shadow economy relationship using new estimates of U.S. state-level shadow economy size, and detailed data on corruption convictions of U.S. public officials from TRACfed. The analysis will consider the corruption-shadow economy relationship as it relates to various types of corruption.

The International Human Rights Law Clinic at the University of California, Berkeley School of Law (Boalt Hall) is conducting research as part of a project assessing the implementation of the TRUST Act (AB 4) in its first year in California and identifying the advances and gaps in implementation around the state. The results of the research as well as recommendations more thorough and consistent implementation will be documented in a final report, published just after the one-year anniversary of implementation.

The TRUST Act (AB 4), signed in to law on October 5, 2013, has made California a leader in pushing back against the wave of enforcement by narrowing the categories of immigrants that can be held by local law enforcement agencies at ICE's request. The TRUST Act, which went in to force on January 1 2014, sets a minimum standard across the state to limit immigration "hold" requests in local jails but is being unevenly implemented at the local level.

The project team is in the midst of compiling and analyzing data collected by project partners regarding violations and also conducting research regarding local law enforcement policies and practices, particularly in those counties where little data has been collected. Research includes interviews with advocates, law enforcement officials, immigration officials, and formerly detained non-citizens. The team has submitted public records requests to county sheriffs across the state and is using the limited, publically available data from federal agencies. Upon completion of research, a report summarizing findings, identifying patterns of both the lack of implementation and best practices, and making recommendations to strengthen implementation will be released.

Central to the emerging debate over expanding access to counsel for poor immigrants facing possible deportation are assumptions about the availability of counsel and the relationship of attorney representation to case resolution. Yet, at the national level, surprisingly little is known about the current level of immigrant representation, the institutions that are providing immigration representation, and the impact of representation on case outcomes and court efficiency. This Article, which draws on an extensive dataset of over 1.2 million deportation cases decided by immigration judges between 2007 and 2012, presents the results of the first national study of attorney representation in United States immigration courts. The study’s findings amplify recent regional efforts to study patterns in access to counsel, while adding a valuable national perspective to inform the ongoing conversation regarding how best to increase attorney involvement in immigration court.

The U.S. Congress passed the Sarbanes-Oxley Act in 2002 following a series of high-profile corporate scandals, thereby raising the visibility of corporate fraud as a serious ethical and legal issue. In this paper, I examine the geographical distribution of corporate fraud cases in the United States, and analyze the lead charges brought by prosecutors in order to determine the most frequent types of corporate fraud observed across states. I also investigate the impact of state-level labor and employment laws on the frequency of corporate fraud convictions in an effort to provide insights on the kinds of policies that are most likely to be effective in deterring fraud.

The aim of the project is to understand the social production, use and impact of terrorism 'expertise'; how and why it differs across selected domains and to assess why 'evidence based' approaches are not always the most intensively utilised. The specific objectives of the project are:

To determine and analyse the processes of knowledge production particular to this field of knowledge;
To determine the funding regime which supports the production of 'terrorism' knowledge in the public sector and in the world of think tanks and the private sector;
To examine how knowledge claims around terrorism are subject to scrutiny and forms of peer review in a variety of domains;
To examine how academic and government knowledge on terrorism is communicated to policy-relevant, academic and public audiences (what are the processes of dissemination particular to this field of knowledge);
To examine and ascertain the relationship between counter-terrorism practice (intelligence gathering, evidence-based policy and strategy etc.) and academic knowledge generation and dissemination;
To produce guidance for both policymakers and media professionals on the use of terrorism knowledge.

This project is aimed at describing and analyzing prosecutorial charging behavior using TRAC charge data. In particular, the project will explore the types of charges that prosecutors bring, and the timeline in which prosecutors charge mandatory minimums, and bring superseding indictments. I will also explore whether prosecutorial charging strategies differ across crime types, defendant types, and sentencing judges using combined data on charging and sentencing.

Since 2010, the Obama Administration has attempted to encourage the use of prosecutorial discretion by Immigration and Customs Enforcement (ICE), the executive agency in the Department of Homeland Security (DHS) charged with enforcement of immigration laws. This effort has been widely decried as ineffectual. Statistics reflect a disappointing number of cases closed pursuant to the program, and advocates and scholars have reported widely on cases that demonstrate the agency's failure to embrace its stated goals.

My research seeks to document the key role of criminal convictions in ICE agents' decision making process, which is a central reason for the failure of the prosecutorial discretion initiative to take hold. I will demonstrate in two ways the agency's heavy reliance on criminal convictions as a key metric for case processing determinations. The first is a case study of a client of the University of Arizona's Immigration Clinic, who appeared to be a quintessential recipient of prosecutorial discretion, and yet ICE subjected her to two years of detention and aggressive prosecutorial tactics due to a single conviction for which she received probation only. The second is an analysis of data obtained through TRAC of all cases closed pursuant to prosecutorial discretion, which I will analyze to determine the proportion of cases closed that involve criminal convictions of any kind.

Building on this data, I will argue that one key reason that prosecutorial discretion has not taken hold within ICE is the failure of the executive branch to adequately account for the powerful role of agency culture. Prosecutorial discretion, particularly of the kind that is intended by the ICE directives, requires agents to draw on internal, often informal, forms of regulation to decide whether to proceed on an individual case. In ICE's case, all these internal cues point in the direction of enforcement when an immigrant has a criminal conviction, regardless of its seriousness. This heavy reliance on convictions directly contradicts the prosecutorial discretion initiative, which was undertaken in part to respond to concerns that ICE was focusing its enforcement efforts on immigrants with non-serious, misdemeanor convictions.

When decision makers review many cases, how might they be biased by the desire to be fair or appear fair? A large body of work suggests social preferences are driven, in part, through social audience or self-image. Using four datasets, we test for cold-hand decision-making by asylum officers and other decision makers and explore factors that ameliorate or exacerbate this potential cognitive bias. Namely, we test for behavior consistent with the belief that if previous decisions are called in one direction by the decision maker, future decisions have a greater chance of being called in the opposite direction. We also explore manipulation in terms of the overall sequence of favorable vs. unfavorable decisions.

We test the claim that purely personal, non-normative concerns ("what the judge ate for breakfast") may influence judicial decision-making. We link asylum grant rates in US immigration courts to data on the success of the local professional or college sports team on the night preceding the decision. A critical assumption for our test is that sports outcomes are orthogonal to the merits of (asylum) cases. Even if hearing schedules took into account sports schedules (e.g., "big games"), we could eliminate any resulting bias by looking not at sports outcomes per se but deviations from expected outcomes. Another concern is if asylum litigants' behavior changes as a result of sport games outcomes. We will therefore link judges to sports teams of their native or university towns, which is more likely to be exogenous to asylum litigants' behavior.

Abstract not yet available.

We show that isolated capital cities are robustly associated with greater levels of corruption across US states, in line with the view that this isolation reduces accountability, and in contrast with the alternative hypothesis that it might forestall political capture. We then provide direct evidence that the spatial distribution of population relative to the capital affects different accountability mechanisms over state politics: newspaper coverage, voter knowledge and information, and turnout. We also find evidence against the capture hypothesis: isolated capitals are associated with more money in state-level campaigns. Finally, we show that isolation is linked with worse public good provision.

This research examines the impact of tax enforcement on the incidence of stock option backdating. Consistent with the theoretical prediction that tax authority enforcement can operate as a valuable monitoring tool by narrowing the scope for managerial entrenchment, I find robust evidence that the incidence of stock option backdating is lower when firms are more likely to be subject to IRS audits. These results reinforce calls in the public policy discourse for institutions that protect investors by curtailing companies' "degrees of freedom" to engage in corporate misbehaviour.

Prior research finds that federal expenditures on state-level programs tend to exacerbate public corruption by increasing opportunities for fraudulent appropriation. In this study, we plan to examine the relation between public corruption and various categories of government expenditures. We are interested in both the impact of expenditures on corruption, and the potential for corruption to alter the pattern of expenditures.

One area of particular interest is the relation between federal expenditures for disaster relief and corruption. Most of the disaster relief expenditures are either distributed by FEMA directly to affected individuals, or distributed by funneling funds through state and local government agencies. Several types of assistance could be made available to individuals, businesses and the public after a major disaster. For example, state and local governments are eligible to receive financial and technical assistance from FEMA in the form of grants for emergency protective measures, repair, or restoration of damaged facilities. Qualified individuals and businesses are eligible to receive assistance from the SBA in the form of low interest loans, and households may receive food stamps through the Food and Nutrition Service of the USDA.

We obtain state-level data on the number of disaster declarations and government expenditures from the Public Entity Risk Institute and the U.S. Census. Our measures of corruption include the number of prosecutions and convictions per capita for corruption at the federal, state and local levels, and the number of prosecutions and convictions per capita for corruption at the federal agency-level. To examine whether disaster aid expenditures have any impact on public corruption we estimate ordinary least squares regressions that include state and year fixed effects. We also control for other state characteristics that are known to affect corruption. The results of this study should increase our understanding of the relation between government spending and corruption.

Read the paper: Do State Campaign Finance Reports Reduce Public Corruption?

Judges play a central role in the American legal system-more so than in any other legal system. But the behavior of judges, and in particular the determinants of their decisions, are not well understood by lawyers, including law professors; in part this is because judges in our system are permitted to be (and are) quite secretive. Beginning more than half a century ago but accelerating in recent decades, social scientists-political scientists in particular, but also economists and psychologists, and some academic lawyers trained or interested in social science-have used increasingly sophisticated theoretical concepts and quantitative tools to place the study of judicial behavior on a genuinely scientific basis. Yet no single book provides a comprehensive, up-to-date overview of the new learning about judicial behavior, and it is that gap that we hope to fill with a combination of survey and original research.

This book presents our findings with respect to judicial behavior in the three principal tiers of the federal judiciary-the U.S. Supreme Court, the federal courts of appeals, and the federal district courts. We propose and test a variety of hypotheses concerning influences on judicial voting and on other aspects of judicial behavior. A major focus, of course, is on the relative weight of ideology and legalistic analysis in judicial decisionmaking in the different tiers. But we also measure more mundane, though we think very important, influences on judicial behavior, such as effort aversion (which includes but is not exhausted in leisure preference), that are predicted by our labor-economics model of judicial incentives and constraints.

Utilizing TRAC's dataset on sentencing behavior in federal district courts, we will determine whether federal district judges interested in promotion to a higher court alter their sentencing behavior to improve their promotion prospects. Specifically, we would benefit from access to a dataset which lists the following attributes for each case: the district, the district judge's name, the president who appointed the judge, the year the judge was appointed, the institution from which the judge received a law degree, the judge's age at the time of sentencing, the defendant's race, the defendant's gender, the program category, the lead investigate agencies, the lead charge, whether the sentence followed a bench trial or jury trial, whether the defendant entered a guilty plea, the sentencing date, and the sentence. This analysis could be paired with our findings about federal courts of appeals justices seeking promotion to the U.S. Supreme Court.

Sentencing disparities by race, gender, education, and socioeconomic status are prevalent in the federal criminal justice system. Black defendants are sentenced to almost five months longer in prison than white defendants with identical demographic traits, criminal history and offense characteristics. Male defendants receive over five months longer in prison compared to similar female defendants, and defendants with lower educational attainment and income receive significantly longer sentences than otherwise similar offenders.

These large disparities in federal criminal sentencing prompted calls for sentencing reform in the early 1970s. In response to concerns that judges were introducing unwarranted disparities in sentencing, Congress adopted the United States Sentencing Guidelines under the Sentencing Reform Act (SRA) of 1984. After almost two decades of mandatory Guidelines sentencing, the Guidelines were struck down in United States v. Booker, 543 U.S. 220 (2005). As a result, Booker greatly increased the degree of judicial discretion afforded to judges.

This paper first estimates the impact of increased judicial discretion via Booker on racial disparities in federal sentencing using comprehensive data on the universe of defendants sentenced between 1994-2009. Next, I examine the sources of increasing disparities after Booker by studying how different types of judges have responded to increased judicial discretion. Matching three data sources, I construct a dataset of over 400,000 criminal defendants linked to sentencing judge from fiscal years 2000-2009. Given that cases are randomly assigned to judges within a district court, any difference in sentencing practices across judges can be attributable to the judge differences, rather than case composition. Exploiting the random assignment of cases to judges in this dataset, I specifically explore the impact of experience under a mandatory Guidelines regime on sentencing practices to test the hypothesis that judges appointed after Booker develop different sentencing practices compared to their colleagues.

I conclude by considering the impact of judicial discretion on other actors in the criminal justice system. I find evidence that increased judicial discretion via Booker changes the prosecutorial treatment of statutory mandatory minimums, which Booker left intact, consistent with prosecutors attempting to rein in judicial discretion. Future work on the interactions of actors at various stages in the criminal process is critical to a thorough exploration of disparities in the federal criminal justice system.

See discussion paper for initial findings from this research.

Facilities to detain immigrants have been built across the United States. Local communities clamor for them in hopes of generating revenues for local units of government and jobs for local residents. To date, no study has examined whether or not these purported benefits materialize. It is anticipated that this study will provide no evidence that ICE facilities promote local employment growth. Digging deeper, analyses will be undertaken to determine if ICE facilities are doing more harm than good (i.e., impeding job creation.

Opinion polls confirm that public mistrust of government and concern about political corruption are on the increase. Nevertheless, very little is known about the frequency of different types of corrupt activities or even what types of public officials are more prone to corruption. We analyze lead charges in public corruption cases over 25 years in order to describe the types of corruption observed more or less frequently in practice. This study will better inform policy makers regarding what kinds of policy responses may be most effective in reducing political corruption.

The public debate over state political reforms such as legislative term limits, voter identification or campaign finance and lobbying reforms, often includes strong assertions that institutional reforms will reduce political corruption and public cynicism regarding the integrity of democracy. However, such claims have not been well-tested, in part for want of appropriate measures of political corruption. This study is the first to evaluate the within state effects of institutional reforms on political corruption and in turn, the within state effects of corruption on political behavior.

We measure state political corruption by public official prosecutions and convictions per state full-time equivalent employment; we then estimate a fixed effects model to identify the impact of state political institutions on corruption, controlling for other state characteristics. We also use these measures of state-level corruption to examine whether public corruption has any impact on individual-level political behavior such as trust in government and voter turnout. We analyze a mixed-level model with state fixed effects to identify the impact of state-level corruption on individual-level trust in government, controlling for other relevant individual and state-level characteristics.

While the fixed effects that we employ ameliorates concerns about simultaneity bias in these regression models by sweeping out time-invariant unobservable determinants of corruption, we also check time trends in corruption before and after episodes as a further check for confounding effects form time-varying unobservables. We also test whether particularly prominent episodes of corruption precede reform efforts in the states. Finally, we will examine the determinants of state-level corruption itself in order to explore whether there exist appropriate instrumental variables that help predict state-level corruption. Potential instruments include state-level proxies for the historical development of organized crime, such as the timing of prohibition laws and murder rates for young men during prohibition.

In 2006, Congress allowed videoconferencing for conducting immigration removal hearings. As a result of this legislative change, immigration judges around the country now conduct immigration hearings in which the respondent appears in the courtroom via video screen. This practice is particularly prevalent in cases in which the respondent is detained. In fact, in many immigration courtrooms around the country, hearings of detained respondents are now held exclusively by videoconference. With the corresponding increase in the use of detention to house individuals petitioning for immigration relief, the practice of videoconferenced immigration hearings has expanded greatly over the past few years.

The benefits of videoconferencing are often claimed to be improved safety, judicial efficiency, and reduced transportation costs. However, the effects of this shift in the judicial process on outcomes of immigration removal hearings have not been systematically analyzed. This project will draw on TRAC's extensive databases of immigration proceedings to explore the effects of videoconferencing technology. In particular, this research seeks to understand how videoconferencing is used across multiple jurisdictions and what, if any, effects videoconferenced (as opposed to in-person) participation has on substantive outcomes.

Prosecutors possess wide discretion to determine which individuals receive criminal punishments and which do not. The analysis proposed here will bring more closely together the conception of prosecutors as labor market actors and the decisions prosecutors make in criminal adjudications. It will do so by providing a more thorough empirical examination of the decisions of prosecutors and the duration of their careers as prosecutors.

The research will make two primary contributions. First, it will provide a statistical portrait of the typical federal prosecutor or Assistant U.S. Attorney. Surprisingly little is known about the careers of line prosecutors in U.S. Attorneys' Offices, and for that reason, basic facts about their work will be an advance. In addition, the research will pay close attention to how the professional experience of the "average prosecutor" varies across district. Second, the research will more closely examine the types of legal work that correlate with turnover in the prosecutor's office.

I plan to examine these relationships with hazard models. Results from this estimation will expand our understanding of the career trajectories of federal prosecutors. Perhaps more importantly, these results may provide a first clue to whether a particular set of extra-legal factors – specifically, the prosecutor's career concerns – influence the types of charges brought and the willingness to take cases to trial.

Information visualization is usually viewed as a neutral means to support analytic reasoning. In recent times, however, designers and artists are joining the emerging field of artistic data visualization to create purposive and original representations of information.

Developed based on TRAC's data collections, this artistic data visualization project explores the symbiotic relationship between creative design and information visualization. It analyzes federal enforcement, demographic and economic information of each state in the U.S. with the aid of TRAC's data tools. Instead of directly representing available information, this project interprets the statistics with an aesthetic oriented approach. It associates each data type with a graphic pattern, and uses the patterns as compositing components to construct 3D computer graphics as profiles of the individual states. The graphics are created with a generative algorithm and rendered by a photorealistic rendering engine. While offering aesthetic satisfaction, the use of patterns in the visual presentation should also help the viewer's analytical reasoning and encourage a contextualized reading of the information.

Some images can be viewed at reddotbluesquare.com.

Political science scholars, as well as legal scholars, are currently researching questions of prosecutorial discretion, specifically why prosecutors make the choices they do. These scholars debate whether the centralizing influence of Department of Justice officials or the decentralizing forces of local political culture play a greater role in influencing prosecutorial agenda-setting in U.S. Attorneys' Offices. Political scientists also are concerned with the causes and effects of immigration policy, investigating such questions as whether changes in macroeconomic conditions or the geographic concentration of immigrants affect public attitudes toward immigrants, as well as whether presence of nativist sentiment among voters yields changes in policy.

We propose to study the prosecution of immigration offenses in the four California U.S. Attorneys' Offices from 1996 to 2007. Our dependent variable (decision by prosecutor whether to decline or prosecute an immigration offense referral) would be compared with Department of Justice prioritization (measured by frequency of mention of immigration prosecution in Attorneys' General speeches), local political culture (measured by region-specific public opinion polls concerning attitudes towards illegal immigrants), economic conditions (measured by region-specific unemployment rates) and Office resources (measured by available FTEs). We hypothesize that local political culture will have a significant influence on the decision making of federal prosecutors—prosecutors will be less likely to emphasize immigration prosecutions in a liberal environment such as San Francisco rather than a conservative environment such as San Diego. We further hypothesize that macroeconomic conditions will not appreciably affect prosecutorial agenda-setting.

Political science scholars, as well as legal scholars, are currently researching questions of prosecutorial discretion, specifically why prosecutors make the choices they do. Currently many social scientists disagree about the extent to which variations in regional political culture shape political outcomes, including prosecutorial decisions, in the United States.

We propose to study federal obscenity prosecutions over time, in order to answer the question of why some federal prosecutors prioritize these cases while others do not. We hypothesize that these prosecutions are more likely to occur in "red states" rather than "blue states", are more likely to occur under Republican rather than Democratic presidential administrations, and are likely to be of a qualitatively different character in conservative political environments. More specifically, we anticipate that prosecutors working for a Democratic administration or working in a "blue state" are more likely to bring obscenity prosecutions for charge stacking purposes whereas prosecutors working for a Republican administration or working in a red state are more likely to bring these prosecutions in order to influence the qualitative nature of sexually-explicit material within their communities. Finally, we expect that the gender of the prosecutor will have no effect on the tendency to bring obscenity prosecutions.

By understanding the relationships between presidential influence, local political culture, and prosecutorial strategies, this project will lead to a deeper understanding not only of prosecutorial behavior, but also of the extent to which differing regions of the United States truly are involved in a "culture war."

List of TRAC Fellows